Many kinds of parenteral drug therapy require continuous drug delivery in preference to single or multiple drug injections. Benefits that accrue from continuous therapy may include, for instance, reduction of toxic or other side effects associated with sharp pulses of drug, significant improvement in the effectiveness of the therapy through the use of smaller amounts of drug, and increased patient comfort. The traditional manner of administering sustained parenteral treatments is via intravenous drip. Intravenous drip treatment is commonplace in a hospital environment, but this treatment mode obviously imposes severe restrictions on the activity of the recipient. As a result, considerable research over the last few years has been devoted to the development of small, portable infusion pumps. A range of devices has appeared, including those with electric or clockwork motors that drive syringe or peristaltic pumps, and others powered by the elastic tension of an inflated balloon, or the vapor pressure of a volatile propellant. Literature incorporated herein by reference describing such pumps are Controlled Release Micropump for Insulin Administration, (M. V. Sefton et al., Ann. Biomed. Eng., Vol. 7, pp. 329-343, 1979), Continuous Intravenous Arabinosyl Cytosine Infusions Delivered by a New Portable Infusion System, (J. Bottino et al., Cancer, Vol. 43, pp. 2197-2201, 1979), or product brochures from Auto-Syringe, Inc., Hooksett, N.H. and Cormed, Inc., Medina, N.Y. These devices are typically strapped onto the wearer, or carried on a belt or in a harness. Also, most are designed to deliver relatively large quantities of fluid and do not effectively dispense small volumes of the order of a few milliliters or less.
An alternative approach that has been exploited to a limited extent is to drive the infusion device osmotically, using a Rose-Nelson pump, activated by imbibition of water or other driving fluid. The principle of the osmotic pump was originally conceived by Rose and Nelson in the 1950's (S. Rose and J. F. Nelson, "A Continuous Long-Term Injector," Austral. J. Exp. Biol. 33, pp. 415-420 (1955)). A Rose-Nelson pump consists of three chambers: a salt chamber containing excess solid salt, a drug chamber, and a water chamber. The salt and water compartments are separated by a rigid membrane permeable to water but impermeable to ionized and hydrated salt ions; the salt and drug chambers are separated by a rubber diaphragm. In operation, water is imbibed osmotically into the salt chamber, causing the rubber diaphragm to expand into the drug chamber and forcing the drug out through the delivery orifice. Depending on the salt used, the osmotic pressure developed by this type of pump is usually between 50 and 200 atmospheres. The pressure required to pump the drug from the device is small in comparison, and hence the drug delivery rate remains constant as long as some excess undissolved salt remains in the salt chamber. In comparison with mechanically-driven devices, Rose-Nelson pumps are small, reliable, and simple and inexpensive to manufacture. U.S. Pat. No. 3,604,417 discloses a modification of the Rose-Nelson pump in which a movable piston replaces the elastic diaphragm separating the drug and salt chamber, and both the drug and salt are loaded into the pump as solutions. U.S. Pat. No. 4,474,048 discloses another modification of the Rose-Nelson principal employing an impermeable elastic wall, and a movable end wall that can be screwed in to deliver a pulse dose of the contained drug at any time during the operation of the pump. U.S. Pat. No. 4,474,575 is a variant of 4,474,048 in which the flow rate of the dispensed agent can be varied by altering the area of semipermeable membrane exposed to the water chamber. U.S. Pat. No. 4,552,651 discloses a pump assembly with a small osmotic pump that can be filled in advance of use with the active agent to be dispensed. The action of this pump is initiated by filling the lower chamber of the housing with a hydrogel. Once the pump is in action, an optional mechanism for delivering pulse doses can be employed. All these osmotic pumps are self-driven and begin to operate as soon as all of the several chambers are filled with their fluid contents and liquid is imbibed across the semipermeable membrane into the salt chamber.
U.S. Pat. No. 4,838,862, commonly owned with the present application and incorporated herein by reference in its entirety, describes a portable osmotic infusion pump that can be filled with the agent to be dispensed, the osmotic salt and the driving medium, and then stored as a complete assembly, ready for activation and use without need for addition of other components. U.S. Pat. No. 4,898,582, also commonly owned with the present application and incorporated herein by reference in its entirety, describes a portable osmotic pump that includes a housing with two side-by-side compartments, where one compartment contains the osmotic salt chamber, and the second compartment contains the imbibing liquid for the pump. The latter two patents describe osmotic pumps that can be filled with all required fluids, including the drugs to be delivered, stored until needed, and then activated very rapidly on demand. They are therefore excellent systems for use as disposable drug infusion devices.
Many limitations of these infusion devices, however, have not yet been addressed or resolved. Assembled infusion pumps contain trapped air spaces in a variety of areas such as the delivery lines, the needle assembly and interior spaces within the body of the device. Trapped dead air spaces impede the onset of infusate flow by attenuating the expansion effect of the motive force of the device. These dead space areas will accept the force developed by the motive power without a concomitant infusate flow. Infusate flow from the pump device will be delayed until all of the communicating air spaces are removed or compressed to the point where they do not affect infusate flow. Another common limitation of some of these systems is that the patient does not have control over activation of the device, or has, at best, only limited control. For example, if a device is activated by ingestion, it will begin to release drug as soon as it comes into contact with internal fluids. Another limitation is that, for many of these systems, the delivery rate of the infusate is not controlled, and, in particular, cannot be controlled by the user. A related problem is that, for many of these devices, the delivery rate of the infusate is controlled by directly regulating the flow of the infusate out of the device, for example, by a valve to control the flow rate. These configurations present problems with sterilization of the device, due to the presence of small compartments and crevices that will come in contact with the infusate that may be difficult for the sterilizing agent to reach. Another problem with flow regulation of the infusate fluid is that shear effects created, for example, by fluid passing through a valve, may lead to degradation of the molecules of drug in solution. For example, proteins or other large molecules are particularly susceptible to shear degradation. Moreover, reliable regulation of these very low flow rates is inherently difficult.
Yet another limitation of some of these devices is that long-term storage of the devices presents problems associated with drug stability and integrity. Many substances, such as drugs, fare poorly when stored, especially when stored in solution. The drug, when stored in a delivery device for a period of time may change or deteriorate chemically and pharmacologically, and may precipitate out of solution. The drug, under these conditions, may also react chemically with other components of the system that diffuse into the drug chamber from various parts of the assembly. This aspect of production, sterilization, and storage of a drug-bearing device is not adequately addressed in available disposable infusion devices and is a problem that therefore limits their use. Another limitation of some of these devices is related to the fact that there is, commonly, only one barrier between the infusate and driving medium, for example, a rubber diaphragm. This configuration creates potential safety problems for the user. Any tear or leak in the wall of the reservoir containing the infusate permits mixing of the infusate with the driving medium, which will result in contamination of the infusate. Such contamination would be potentially harmful to the patient if it happened during use, particularly if the driving fluid were contaminated by bacteria or other harmful substances. Clearly, it is possible to choose a driving medium that would not be harmful to the patient if this accidental mixing were to occur, but the requirement of sterilization of the driving medium, and preservation of the driving medium's sterility during use is yet another obstacle in the creation of a device that is simple and cost-effective to manufacture.
One means for resolving the problems of long-term infusate storage that is described in detail in this application is accomplished by containing the infusate in a removable flexible pouch within the device. In one embodiment of the invention, the pouch could be a part of the device that is filled with the infusate during manufacture, or later, for example by a pharmacist or other person, a short time before the device is used. There are many instances in the patent literature of infusion pumps where the liquid infusate is contained in a separate pouch within the device. U.S. Pat. No. 4,034,756 discloses a small osmotic pump for use in an aqueous environment, such as the gastrointestinal tract, in which the liquid infusate (e.g. a drug solution) is contained in a flexible bag within the device, and the osmotic fluid pressure is exerted directly on the flexible bag to effect infusate delivery. The flexible bag of this patent can be filled with the infusate during pump manufacture, or the bag can be filled with the infusate at a later time. This pump can be activated only by exposure to the aqueous environment, and is therefore limited generally to internal use for drug delivery. The activation means consists of the user swallowing the device or otherwise exposing the device to internal fluids, and rate control is soley a function of the permeability characteristics of the outer semipermeable layer.
U.S. Pat. No. 3,760,805 describes an osmotic dispenser comprised of a water porous housing confining a first flexible bag of relatively impervious material containing an active agent, and a second bag of controlled permeability to moisture containing an osmotic solution. The first and second bags are disposed within the housing such that water permeates from the external environment through the housing and migrates by osmosis into the solution contained in the second bag. The second bag increases in volume, thereby generating mechanical force on the first bag, which mechanical force in turn ejects the active agent out of the device. This pump, designed primarily for ingestion or implantation, depends upon permeation of water from the environment, e.g. gastrointestinal tract, and therefore is unsuitable for subcutaneous infusion.
U.S. Pat. No. 4,201,207 discloses an elastic bladder pump filled with liquid under pressure, which is powered by the elastic tension of the bladder. The bladder can be filled with infusate liquid during or after manufacture. This device also includes a flow control element between the bladder and the catheter fitting, to deliver the liquid infusate to the patient. Because the flow control element operates on the liquid infusate to be delivered, it has some of the attendant problems with shear effects and sterilization as described above. In addition, the bladder must be made of elastic materials, which are inherently permeable. The infusate is therefore vulnerable to absorption of contaminants by diffusion into the infusate from and through the bladder as well as loss of the drug by diffusion through the bladder if it is stored in the bladder for extended periods of time.
U.S. Pat. No. 4,191,181 discloses a liquid infusion pump with a power supply such as a battery and a refillable flexible infusate reservoir. This pump can be activated on demand and has a flow control means that acts directly on the pumping mechanism, which is downstream from the infusate reservoir. Therefore, the rate control means, in this case the pump mechanism, comes in direct contact with the infusate, causing the attendant problems with sterilization and shear effects described above.
U.S. Pat. No. 4,596,575 discloses an implantable liquid infusion pump that is particularly intended for the delivery of insulin. It contains two collapsible reservoirs in rigid housings, and also a mechanical pump that is regulated by an electronic unit control for management of pump activation and flow rate. One of the reservoirs contains the infusate; the space between the outer wall of this reservoir and its rigid housing is filled with the drive liquid. The second reservoir is filled with the drive liquid, and the space between the outer wall of this reservoir and its rigid housing is maintained at subambient pressure. The drive liquid is pumped from the second reservoir into the outer space of the first housing to exert pressure on the first reservoir and thus deliver the liquid infusate. The electronic control unit regulates two valves that restrict the flow rate of the drive liquid; thus the infusate does not come in contact with the valving system. The device is also provided with a separate refill system that may be used to refill the infusate reservoir. This type of refill mechanism, however, presents sterility problems during long-term use, particularly because it is used in an implantable device that cannot be cleaned during use. In addition, if the pouch wall breaks or tears, the infusate is susceptible to contamination from the driving medium, as described above.
Other patents in the literature describe portable infusion pumps that contain flexible pouches containing the infusate fluid that are removable from the device, or that can be loaded separately into the device after manufacture of the main pump assembly. U.S. Pat. No. 4,193,398, for example, discloses an extracorporeal osmotic pump in which the infusate liquid is contained in a flexible pouch that is located within a second pouch containing the osmotic fluid. In one embodiment, the infusate pouch can be removed from the main pump assembly and be replaced. This pump also contains an additional chamber filled with the driving medium for the pump, for example water or a weak osmotic solution, and incorporates rate-control and activation means. The osmotic fluid, however, has the potential to be in direct contact with the outer wall of the infusate pouch. Therefore, diffusion of the osmotic fluid into the infusate, and resulting damage, can occur if the infusate is stored within its pouch for any length of time in its fully assembled state. In addition, if the pouch wall breaks or tears, the infusate is susceptible to contamination from the osmotic fluid, as described above.
U.S. Pat. No. 4,398,908 discloses an infusion pump driven by electromechanical power, in which insulin is stored in a flexible pouch that is removable and replaceable, as is the pumping mechanism described. This pump can be activated on demand and incorporates a rate-control mechanism. The pumping mechanism, which incorporates the activation and rate-control means, is downstream from the insulin reservoir, and therefore comes in direct contact with the insulin infusate. As a result, this design has problems with sterilization of the pumping mechanism during manufacture, and shear effects on insulin during use, as described above. Insulin, being a large biomolecule in solution, is particularly susceptible to such damage.
U.S. Pat. No. 4,525,164 discloses a motor-driven pump with a removable and replaceable arcuated reservoir, in which a piston applies pressure directly to a portion of the reservoir to propel the drug out of the device. This device incorporates a means for activating the pump motor and controlling the pumping rate. Because the piston acts directly on the outer wall of the infusate reservoir, however, there is the possibility of shear effects on the infusate drug. Also, maintaining sterility in the pouch and pumping mechanism is difficult.
Each of these references describes an infusion pump that incorporates a separately loaded pouch or reservoir in a specific configuration, and usually with a specific motive force, yet all have problems that have inhibited their use. These problems include a high cost of manufacture, difficulties in sterilizing the drug chamber and contents, difficulties in maintaining sterility of the device, and problems with stability of the devices after prolonged storage. As a result, there remains a need for a reliable disposable infusion pump that can be loaded with sterile liquid infusate, that can maintain sterility during prolonged storage, and that can be activated on demand to provide the required pattern of delivery of very low infusate flow rates.
The present invention describes an infusion pump assembly that incorporates an infusate pouch that can be manufactured, sterilized, and aseptically loaded with infusate and assembled into a sterile pump assembly or sterilized separately from the other components of the assembly and then loaded into the pump assembly at the appropriate time. The pump is small, light, and convenient for patient use, and may be driven by a variety of motive forces in its different embodiments. The infusate pouch of this invention is simple in design, and therefore would be straightforward to manufacture and sterilize. It can also be made of nonelastic materials and therefore can be constructed using materials that are relatively impervious to invasion by environmental agents such as oxygen, carbon dioxide, or substances that are derived from components of the device. The rate of delivery of the infusate from this pouch is controlled by the expansion of a second pressure-transmitting pouch. Pressure is obtained by filling the second pressure-transmitting pouch with a driving medium at a controlled rate.
Presentation of sterile medication to the patient via infusion devices presents several unique problems in drug stability and sterility. Deposition of a drug solution into a cavity within an infusion device just prior to application requires presterilization of that cavity and delivery of a unit dose of sterile drug to the aforesaid cavity in an aseptic manner. Some methods of sterilization may be suited for some materials but totally unacceptable for other desired components. Many drugs and solutions may be rendered sterile by gamma radiation, but polypropylene components subjected to sterilizing doses of gamma radiation suffer severe radiation induced degradation. Metals may be advantageously sterilized using steam, but many drugs cannot withstand the steam sterilization regime. Separation filling and sterilization of the infusate pouch from the manufacture and treatment of the remainder of the infusion device permits optimization of the sterilizing procedure. For example, the infusate pouch may be manufactured and sterilized under conditions most suited for the materials of pouch construction. Then, at a later date, the sterile pouches may be aseptically filled with unit doses of drug solution that have been presterilized by conventional means; or alternately, the solutions may be sterilized by sterile filtration at the point of filling and the pouches sealed. These sterile dose units may then be assembled into the remainder of the infusion device or stored and shipped separately. Since the contents of the infusate pouch do not come into direct contact with the other components of the pump, it is not essential that the remainder of the device be manufactured and maintained under sterile conditions.
Drugs that have limited lifetimes in solution may be stored in a dry or lyophilized form in a two-compartment embodiment of the infusate pouch. In this pouch, one compartment contains the drug and another contains solvent for the drug in a separate sealed part of the pouch. The seal between the two compartments is broken just prior to use, the contents are mixed in the pouch, the pouch loaded into the infusion device, and the device is activated and attached. In this way, the shelf-life of the drug is dependent solely upon the storage conditions of the drug-containing infusate pouch.